A recent report highlighted the fragmented nature of organizations’ AI strategies, with 28% of IT leaders describing their approach as disjointed. Additionally, 35% of organizations create separate AI strategies for individual functions, and 32% set entirely different goals, leading to hindrances in collaboration and alignment of priorities. The lack of a unified approach not only risks duplicating efforts but also overlooks crucial considerations such as ethics and compliance.
Ethics and compliance were found to be overlooked by IT leaders, with legal/compliance and ethics being considered the least critical factors for AI success. Shockingly, 22% of organizations do not involve legal teams in AI strategy conversations, highlighting a significant blind spot in their approach. Neglecting ethical and compliance considerations can lead to the development of AI models that lack necessary standards, resulting in negative impacts on brand reputation, lost sales, costly fines, and legal battles.
Moreover, the quality of AI models’ results is directly linked to the quality of data they utilize. With data maturity levels remaining low and half of IT leaders admitting a lack of full understanding of IT infrastructure demands throughout the AI lifecycle, the risk of developing ineffective models, including AI hallucinations, increases significantly. The energy-intensive nature of AI models can also contribute to unnecessary increases in data center carbon emissions, impacting a company’s brand and return on capital investment.
In order to navigate the complexities of AI successfully, organizations must take a holistic and well-planned approach. This includes training and tuning models across various platforms, from on-premise to the public cloud, to leverage AI’s potential in transforming data into actionable insights. However, it is crucial for companies to balance their desire to innovate with a thorough understanding of the gaps in the AI lifecycle to avoid capital investments that yield negative returns.
Dr. Eng Lim Goh, Senior Vice President of Data and AI at HPE, emphasized the importance of recognizing and addressing these blind spots for AI success. He highlighted the need for organizations to focus on data maturity and align data management practices with their AI goals. Understanding the computing and networking demands throughout the AI lifecycle, as well as ensuring resource provisioning and infrastructure scalability, is essential for leveraging AI effectively.
In conclusion, organizations must adopt a hybrid approach and modern AI architecture to deliver on the promise of Generation AI. By carefully balancing innovation with a deep understanding of the AI lifecycle, companies can avoid costly mistakes and maximize the potential of AI to drive insights and value across their networks.
Article Source
https://technologymagazine.com/articles/hpe-businesses-must-tackle-blind-spots-in-ai-strategies